- From: Anne van Kesteren <annevk@opera.com>
- Date: Thu, 03 May 2007 11:18:18 +0200
- To: "David Dorward" <david@dorward.me.uk>, www-html@w3.org, public-html@w3.org
On Thu, 03 May 2007 11:11:33 +0200, David Dorward <david@dorward.me.uk> wrote: > On Thu, May 03, 2007 at 08:53:21AM +0200, Anne van Kesteren wrote: >> However, breaking backwards compatibility is only possible when we >> introduce a new format as explained on this list before (and that's not >> the goal of the HTML WG). Hence, we have to make user agent requirements >> that require support for the features we rather see people not using. >> (Those features can still be non-conforming, obviously.) > > No, we don't. HTML 5/4.02/4.1 does not need to include legacy and > proprietary features, a user agent can support both "new HTML" and > "Legacy/Proprietary code" without the latter being specified in the > former. You can't specify parsing of the HTML syntax in such a way, unfortunately. It would also be way more editorial work that isn't really justified in any way. I would you suggest you look more into what's actually required to make this work as opposed to simply stating that it can (and should?) be done that way. > For user agents that want to implement legacy and proprietry features, > a seperate specification can be written. > > This may or may not require switching of parse modes for the markup > based on some versioning information in HTML > 4.01. Implementing yet another parser for HTML is not acceptable. Browsers already have to ship with a parser for HTML and one for XML. No need to add yet another one with its own share of bugs. -- Anne van Kesteren <http://annevankesteren.nl/> <http://www.opera.com/>
Received on Thursday, 3 May 2007 09:18:31 UTC